Making Sense of Sensemaking: An analysis of leadership in times of crisis

TABLE OF CONTENTS

List of Tables ... 1

List of figures ... 2

Abstract ... 3

1. Introduction ... 4

1.1 Research goal and question ... 5

1.2 Academic relevance... 5

1.2 Societal relevance ... 6

1.3 Outline of the research ... 6

2. Theoretical Framework ... 8

2.1 Leadership ... 8

2.2 Decision-making ... 10

2.1.1 Decision-making vigilance ... 11

2.3 Sensemaking ... 12

2.4 Sensemaking strategies ... 14

2.4.1 Guided sensemaking strategy ... 15

2.4.2 Fragmented sensemaking strategy ... 16

2.4.3 Restricted sensemaking strategy ... 16

2.4.4 Minimal sensemaking strategy ... 17

2.5 Causal mechanism: connecting strategies to decision-making vigilance ... 17

3. Methodology ... 20

3.1 Research design ... 20

3.2 Data collection ... 20

3.3 Case selection ... 20

3.4 Operationalization ... 22

3.5 Data analysis ... 27

3.6 Reliability and validity ... 28

4. Analysis ... 30

4.1 Organization A ... 30

4.1.2 Decision-making vigilance organization A ... 35

4.1.3 Sub conclusion organization A ... 37

4.2 Organization B ... 38

4.2.1 Sensemaking strategy organization B ... 38

4.2.2 Decision-making vigilance organization B ... 42

4.2.3 Sub conclusion organization B ... 43

4.3 Organization C ... 44

4.3.1 Sensemaking strategy organization C ... 45

4.3.2 Decision-making vigilance organization C ... 47

4.3.3 Sub conclusion organization C ... 50

4.3 Organization D ... 51

4.3.1 Sensemaking strategy organization D ... 51

4.3.2 Decision-making vigilance organization D ... 55

4.3.3 Sub conclusion organization D ... 56

5. Conclusion ... 58

5.1 Reflection ... 59

5.2 Discussion ... 61

5.3 Recommendations ... 62

5.3.1 Social anchors: a network of external experts ... 62

5.3.2 Building resilience: interorganizational collaboration for joint sensemaking .. 63

5.3.3 Information systems: tools for effective sensemaking ... 63

6. Bibliography ... 67

Appendix A: interview questions Dutch ... 75

1

LIST OF TABLES

TABLE 1–STAGES AND CRITERIA FOR VIGILANT DECISION-MAKING ... 11

TABLE 2–CHARACTERISTICS OF THE ORGANIZATIONS ... 21

TABLE 3–ASSESSMENT OF DECISION-MAKING VIGILANCE ... 22

TABLE 4–OPERATIONALIZATION INDEPENDENT AND DEPENDENT VARIABLE ... 23

TABLE 5–SUMMARY FINDINGS ORGANIZATION A ... 37

TABLE 6–SUMMARY FINDINGS ORGANIZATION B ... 44

TABLE 7–SUMMARY FINDINGS ORGANIZATION C ... 50

TABLE 8–SUMMARY FINDINGS ORGANIZATION D ... 57

2

LIST OFFIGURES

FIGURE 1–STRUCTURE THEORETICAL FRAMEWORK ... 8

FIGURE 2– CAUSAL MECHANISM: THE RELATIONSHIP BETWEEN SENSEMAKING STRATEGY AND DECISION -MAKING VIGILANCE ... 19

FIGURE 3–RESEARCH SCOPE ... 22

FIGURE 4–PROCESS OF ANALYSIS ... 30

3

A

While the importance of sensemaking during crises has often been discussed in the academic literature, empirical research specifically examining the role of crisis managers in such a process appears to be limited. Studies that do touch upon the subject mostly focus on leaders in high-reliability occupations, making case studies in the private sector even more scarce. All the while, crisis managers in business organizations are increasingly facing the challenges of a fast-paced and interconnected society. Confronted with extreme situations, they are not only expected to recognize that something out of the ordinary is occurring, but they are also tasked with engaging others in developing a collective understanding. Therefore, to gain new insights on how exactly crisis managers involve themselves in such sensemaking processes, this study conducted a comparative case study analysis. In doing so, the sensemaking strategies of four crisis managers were identified using a typology developed by Maitlis (2005). Subsequently, to determine how these strategies influenced the capabilities of their organizations, the vigilance of the decision-making process was analyzed. In doing so, seven procedural criteria proposed by Janis and Mann (1977) were assessed. The analysis indicated that when crisis managers exerted a high level of control in the sensemaking process, which was the case in a guided and restricted strategy, their organizations had greater capacity to vigilantly gather, interpret and assimilate information and take decisive actions. Therefore, this result not only opens avenues for future research on such sensemaking strategies, but it also creates an incentive to develop methods that actively improve the sensemaking skills of crisis managers.

4

1.

I

Today, our highly interdependent and volatile world requires organizational leaders to act decisively when facing a crisis. Therefore, when the survival of an organization is at risk, a well-developed skillset is needed to deal with complexity and minimize the impact of an urgent threat (Boin & Renaud, 2013). However, even for experienced leaders, the dynamics of a crisis can make it rather difficult to distill meaningful signals from the 'loud noise’. While attempting to figure out what is going on and why, leaders are often bombarded with a stream of conflicting data on ill-defined events and problems (Combe & Carrington, 2015: 308-309). At the same time, not all individuals within the organization are likely to think or act the same way. Team members will see various parts of the disruptive event as important, depending on their position, values and past experience (Choo, 2007). As a result, they can interpret the same crisis as entirely different (Jong et al., 2016). Arriving at a collective understanding of the scope, characteristics and effects of an evolving threat can then become a time-consuming task (Boin et. al., 2013: 82). Yet, the ability to synthesize the vast amount of circulating information is regarded as highly important to not only develop a common frame, but also to organize a coherent response operation (Wolbers & Boersma, 2013). The organizational process that assists in developing such a common understanding is called sensemaking.

Sensemaking is the process through which individuals can obtain a grip on a disruptive situation. After all, to determine how one should act during a crisis, it is essential to first understand the crisis (Weick, 1995). What has happened? Why is it important? How will it impact the organization? How can the situation be stabilized? All these questions need to be answered to understand the complexity and make the events more manageable (Caughron et. al, 2001). To do so, leaders can extract and interpret important signals, develop plausible explanations and construct a sensible environment for others (Weick, 1995). But, acquiring such skills does have its requirements. Generating an accurate view of a critical situation calls for sensemaking awareness as well as perpetual training. The more experience individuals have, the larger their sensemaking repertoire is (Dekrey & Portugal, 2014: 58). However, as most organizations are not expected to encounter extreme events on a recurrent basis, crisis management, let alone sensemaking, is often not a major point of concern (Boin, 2013). Contrary to most governmental institutions, where crises are expected and trained for, leaders in the private sector might lack the capacity or experience to properly deal with extreme events. In such instances, methods to structurally stimulate sensemaking under high pressure could, intentionally or unintentionally, be absent.

5

1.1

R

Now, exactly how concerning should such an absence be? The question arises whether a leader’s active involvement in the sensemaking process will improve the capabilities of his or her organization during a crisis. To gain new insights on this topic, this thesis will conduct a comparative case study among four major private sector organizations. In doing so, the sensemaking strategies of their leaders will be related to the vigilance of their decision-making process. Being ‘vigilant’ as an organization demands a careful and thorough survey of the objectives, an extensive search for and assimilation of relevant information and the development of a contingency plan once an option has been selected (Janis & Mann, 1977). Based on the sensemaking discourse, it is expected that the vigilance of the decision-making process will be influenced by the various ways that organizational leaders ‘make sense’ of a crisis. After all, by structuring the unknown, sensemaking not only helps leaders to have a better grasp of what is going on, but it also enables them to look for additional information, test hypotheses and reduce uncertainty (Phillips & Battaglia. 2003: 2). At the same time, their organizations are stimulated to talk about what is occurring, to develop different interpretations and, most importantly, to make sensible decisions (Lyhne, 2010). Therefore, by analyzing the vigilance of the decision-making process, the goal of this research is establish which strategy is most conducive during a crisis context. Based on these insights, the following research question will be answered:

1.2

A

Although there is extensive sensemaking literature explaining why organizations find it difficult to process and share information under conditions of stress and uncertainty (Reason, 1990; Kahneman, 2011; Coates, 2012), there seems to be limited research specifically focusing on leaders dealing with critical events. As a result, current studies mostly analyze sensemaking through an organizational lens, by emphasizing the collective nature of the process. However, such a gap in the literature overlooks the fact that a leaders’ perspective can be of vital importance when organizations undergo a crisis. They are often they key figures that people look to for answers and to make sense out of what is happening (Weick et al., 2005). In a way, it could be argued that it is during crises that effective leadership might actually matter most. Therefore, to add to a deeper understanding of a leaders’ role in a time-sensitive and ambiguous context, this thesis will adopt a theory elaboration approach on sensemaking.

Furthermore, although scholars have extensively researched sensemaking and decision-making processes by individuals and organizations, limited research explicitly discusses how they link together.

To what extent does the sensemaking strategy used by leaders during crises influence the decision-making vigilance of their organizations?

6 At first, the two might ‘’seem to be an odd pair of terms to reconcile’’ (Boland, 2008: 55) as decision-making is mostly concerned with evaluating different courses of action and sensedecision-making with structuring the unknown. Nevertheless, sensemaking is what allows individuals to deal with ambiguity and uncertainty by searching for rational accounts. After all, to call events a problem, is in essence the outcome of sensemaking (Muhren, 2011). These accounts then enable the decisions and actions needed to clarify what is happening (Maitlis, 2005: 21). Therefore, as both processes operate simultaneously, the goal of this research is to better understand how these processes can influence and complement each other.

1.2

S

Besides its academic importance, this thesis will also have a societal value for leaders in the private sector. Leaders in a high-reliability organizational context, such as emergency services, military units and fire departments, regularly have to face the challenge of making sense of a rapidly changing and ambiguous context (Baran & Scott, 2010: 42). For these organizations, operating in a high threat environment is part of their daily tasks and often considered ‘business as usual’. Consequently, much emphasis is placed on developing proper procedures and training programs which, consciously or unconsciously, include sensemaking (Boin, 2003). Inevitably, the majority of the scholarship on sensemaking during critical events has also focused on cases dealt by the public sector (Landgren, 2005; Dowding, 2016; de Graaff et al., 2016) . It goes without saying that these academic contributions have produced valuable knowledge on sensemaking as they provide rich and elaborate insights. Nevertheless, considerable work remains in the private sector where a crisis can have equally disruptive consequences. After all, as businesses grown, so do their complexities (Schwandt, 2000). To illustrate, the COT Institute for Security and Crisis Management conducted a survey among Dutch industry leaders, stating that in 2017 63% of the organizations felt that their vulnerability to crises was increasing and one in five had already experienced a crisis in the past. For such organizations, a crisis can produce a fundamental threat to the stability of their operational and information systems, a questioning of their core assumptions and beliefs, and ultimately their survival (Seegers & Ulmer, 2002: 126). Practical knowledge is therefore needed to determine whether certain sensemaking strategies can improve an organizations’ capabilities during such challenging times. As a result, crisis managers can be empowered with a well-developed toolbox to promptly identify and respond to alarming developments.

1.3

O

To start, Chapter 2 will introduce several important concepts. Leadership, decision-making and sensemaking will be discussed to provide a comprehensive theoretical background for the analysis. This section will also present the four expectations that will be researched. Next, in Chapter 3 the methodology will be presented by focusing on the research design and case selection. Also, methods

7 and the techniques used to gather and analyze qualitative data will be described. Furthermore, this section will elaborate on the limitations produced by the chosen data gathering and exploitation methods. Thereafter, the switch from theory to practice will be made in Chapter 4, the analysis. Four case studies in the field of crisis management will be analyzed to research the relationship between sensemaking and decision-making. In doing so, interviews will be conducted with organizational leaders, crisis managers in specific, to determine how their sensemaking strategy might relate to vigilance of the decision-making process during crises. A framework for sensemaking will be used to determine whether crisis managers use guided, restricted, fragmented or minimal strategies to deal with a crisis. Subsequently, the vigilance of the decision-making process will be determined based on seven criteria that increase the probability of a correct process. Lastly, Chapter 5 will summarize the findings and reflect on the limitations of the research. Based on these findings, recommendations will be formulated for crisis manager in the private sector.

8

2.

T

F

To answer the proposed research question, several concepts need to be clarified. Therefore, this section will focus on describing the academic discourse on leadership, decision-making and sensemaking. In view of the research objective, all concepts will be discussed in the context of crises. This thesis regards the presence of a threat, uncertainty and urgency as the three central characteristics for such a context. As a result, the following definition will be used:

After providing a theoretical background, this chapter will also elaborate on the four expectations that will be researched throughout the analysis as well as the causal mechanism explaining the predicted relation. The structure of this chapter is illustrated in Figure 1.

.

Figure 1 Structure theoretical framework

2.1

L

A wealth of research has been devoted to the concept of leadership. However, no universally accepted definition has been formulated as leadership is commonly regarded as broad and highly complex concept. Some have preferred a conceptualization in which leadership is regarded as a process of influencing the activities of a specific group whereas others have emphasized either its goal-oriented nature or its ability to mobilize others to struggle for shared aspirations (Kouzes and Posner, 1995). However, despite the varying views on its definition, researchers seem to have developed a common belief that leadership is a flexible and developmental process, in which each new piece of research builds on and seldom completely disregards that which was derived before it (Nawaz and Khan et al.,

Cri si s m anag em ent Leadership

Decision-making Decision-making _vigilance

Sensemaking Sensemaking _strategies

Guided Fragmented

Restricted Minimal

A crisis is ‘’a serious threat to the basic structures or the fundamental values and norms of a social system, which-under time pressure and highly uncertain circumstances necessitates making critical decisions” (Rosenthal et al., 1989: 10)

9 2016: 1). For the aim of this thesis, the concept leadership and its characteristics will be discussed in the context of a crisis.

Naturally, leaders play an important role in times of crisis. They are expected to think and problem solve in an ambiguous situation involving stress and time pressure. At the same time, leadership means interacting with others to formulate a coherent view of the situation at hand (Combs & Carrington, 2015: 207). Effective communication is therefore required to develop cooperating response teams and garner support for the planning process. Without such clear leadership, there is the risk that the crisis response will not be properly coordinated, causing individual groups to rigidly follow their recovery plans without taking the bigger picture into consideration (Kielkowski, 2013: 62). Therefore, to stimulate flexibility, being able to develop tailored responses might be just as important as a carefully thought-out plan. Leaders can encourage such creative solutions by being open to new approaches and opinions while searching for the appropriate response. Besides internal communication, staying in contact with external stakeholders is also of utmost importance. Leaders and organizations need to recognize that a broad number of their stakeholders, including customers, suppliers, competitors, and other members of their environment, can be affected during a crisis (Stephens et. al., 2005: 394). Therefore, next to setting the direction within the organization, re-establishing confidence among stakeholders should also be a central goal (Lucero et. al., 2009). Effective crisis leadership then means sharing accurate and consistent messages to provide confidence that their interests are indeed part of the decision-making process (Kielkowski, 2013: 64). Overall, it can be argued that successful leadership during a crisis requires the adoption of a complex set of competencies, to guide the organization towards successful recovery (Bolman & Deal, 1997).

Next to organizing a proper response, crises also demand leaders to make difficult decisions. Gathering many approaches and opinions has proven to be essential but at the end of the day leaders have to determine which way to go forward. Especially during crises, leaders are expected to have the formal or informal authority to act decisively based on the available information. To do so, they must try to identify decisions that are critical for the quality of the crisis response (Boin, 2006: 64). As a result, it is argued that the most effective leaders involve themselves rather selectively when it comes to crisis decision-making (ibid.). Nevertheless, leading during a crisis takes much more than taking command and having proper communication strategies. A successful response results from proven leadership developed through an ongoing team effort in planning long before the event, coordination during the actual emergency, and careful post-event review (Kielkowski, 2013). By being aware of these competencies and stimulating training sessions, organizations in the face of adversity can ultimately evolve in a renewed and improved way (Sutcliffe & Vogus, 2003). However, various academics (Wooten & James, 2006; Shaw & Harald, 2004; Burnett, 2002) have argued that, even though most leaders are aware of the negative consequences of a crisis, their on-the-job learning experience and

10 formal training does not properly prepare them for crisis management. As a result, crucial leadership tasks may take a back seat to managing the seemingly more pressing matters (Wooten & James, 2008: 353). However, as it is not within the scope of this thesis to discuss the myriad of crisis leadership competencies, such as early recognition, meaning-making and learning (Boin et al., 2016), this framework will zoom in on two specific elements: sensemaking and decision-making.

2.2

D

-

The ability to reach a good decision in the midst of a crisis is both extraordinarily important as extraordinarily difficult (Sniezik et. al, 2002: 1438). To determine the best choice, organizations have to generate response options, develop evaluation criteria, make recommendations and take targeted actions (Hadley et. al, 2009: 5). All the while, the decision-making process is challenged by the three features of a crisis: time pressure, severe threat and high uncertainty. As a result, crisis decision-making seems to demand flexibility and ad hoc adaptations as ‘’no serious threat can ever be dealt with in routine-like manner’’ (‘t Hart, 1993: 14).

Unsurprisingly, academics have varying views on what these adaptations should be. In the literature, a distinction can be made between rational and non-rational models. Proponents of the rational actor model argue that decision-making means coming up with the ideal solution to the crisis before implementing an action. As a result, this is the domain in which decision-makers have at their disposal the maximum amount of time and information to offer complete certainty about the desirable course of action (Rosenthal & ‘t Hart, 1991: 350). However, others have argued that such a model is inadequate. Proponents of a bounded rationality model argue that, generally, individuals make decisions based upon unavoidable constrains such as their limited thinking capacity and time. Consequently, rather than a rational outcome, individuals often seek satisfactory solutions. The garbage can model on the other hand focuses on the idea that decision-makers act on their gut feelings (Arthur, 1994). As a result, decisions are chance occurrences, produced by several independent streams within an organization. However, despite their differences, both bounded rationality and the garbage can model emphasize that complete rationality is unattainable as, in reality, information is difficult to assimilate, costly to acquire or simply unavailable (Oneal, 1988).

Besides discussing its rationality, scholars have also debated whether decision-making in crisis management should be centralized or decentralized. Hermann (1963) for instance argues that organizations tend to include fewer people in their crisis decision-making process to be more effective. The argument for such a choice is that crises simply ask for ad hoc adaptations of the regular bureaucratic structures and culture (‘t Hart et. al, 1993). The dominant decision structure during such times should therefore be a small group. Dutton and Jackson (1987) support this centralization idea, arguing that it could provide the required higher degree of autonomy. More filters by subunits would

11 mean more chances for delay (Peters et al., 2008). However, other scholars have objected that centralized decision-making should be regarded as the ideal form. Centralization could also put intense pressures on decision-makers that take part in the crisis groups. Besides, group dynamics may increase the danger of systematic exclusion of other important stakeholders. This could ultimately result in the group think phenomenon, through which problems are avoided to maintain a group consensus (Janis, 1972). Lastly, another issue focuses on the possibility of a data overload. As most crises produce an explosion of communications and data, a lack of adequate staffing or clear information monitoring might lead to a limited oversight of the situation (ibid.).

As illustrated, a lively debate has taken place around the concept of crisis decision-making. Although both sides have proposed convincing arguments on the topic of rationality or centralization, this thesis assumes a more pragmatic stance. Ultimately, it’s the characteristics of the crisis and the structure of the organization that determine which insights and strategies might produce the best response. Based on this notion, it is also assumed that the success of a decision-making process should be determined using a set of criteria that focus on the vigilance of its processes. These criteria will be discussed below.

2.1.1DECISION-MAKING VIGILANCE

Janis and Mann (1977) extracted from the academic literature a guideline for the ‘ideal’ decision-making procedure (Fandt et. al., 1993: 129). As the intention of this study is to understand how and why decisions are made during crises, the focus will also be on the context of decision-making rather than the result. To analyze this context, Janis and Mann (1977) present a model that makes a distinction between vigilant and hypervigilant decision-making patterns. Being vigilant is described as having a solid process of decision-making in which the decision-makers canvas an array of alternatives, search painstakingly for relevant information, assimilate information in an unbiased manner, and appraise alternatives carefully before making a choice (Mann, 1982: 73). It is argued that a focus on such an analytical pattern will be beneficial as the quality of a decision is contingent upon the quality of the discussions preceding the actual choice (‘t Hart, 1991: 268). Ultimately, vigilance minimizes the chance of regret that certain decisions were made (ibid.). The four stages and seven criteria, as formulated by Janis and Mann (1977), are as follows:

Decision-making phase Criteria I. Assessment of

the situation

- Decision-makers appraise the situation and determine the full range of objectives to be fulfilled;

II. Surveying of alternatives

- Decision-makers thoroughly canvas a wide range of alternative courses of action;

12

- Decision-makers carefully consider whatever is known about the costs and risks of negative consequences, as well as the positive consequences, that could flow from each alternative;

- Decision-makers intensively search for new information relevant to further evaluate the alternatives;

III. Weighing of alternatives

- Decision-makers correctly assimilate and take account of new information or expert judgement to which they are exposed, even when the information or judgement does not support the course of action they initially preferred;

- Decision-makers reexamine all the possible consequences of all known alternatives, including those originally regarded as unacceptable, before making a final choice

IV. Preparation and monitoring

- Decision-makers make detailed provisions for implementing or executing the chosen course of action, with special attention to contingency plans that might be required if various known risks were to materialize (Jenkins, 1985: 222).

Table 1. Stages and criteria for vigilant decision-making (Janis & Mann, 1977)

However, conditions such as an unexpected threat or time pressure may give rise to hypervigilant behavior. In contrast to a vigilant decision-making process, a hypervigilant pattern is characterized by ‘’(a) a nonsystematic or selective information search, (b) consideration of limited alternatives, (c) rapid evaluation of data, and (d) selection of a solution without extensive review or reappraisal’’ (Johnston et al., 1997: 614). Although hypervigilance might save time and effort, it also represents a disorganized and impulsive method of decision-making. As a result, decision-makers will fail to adequately execute the cognitive tasks that are crucial for arriving at stable decisions (Janis, 1982: 73). Therefore, by meeting every criterion more adequately, the chances for gross miscalculations can be lowered. If, however, fewer criteria are met, the likelihood increases that an organization undergoes unanticipated setbacks and reaches poor decisions (ibid.).

2.3

S

The management of a crisis is a complex and challenging task (Boin et al., 2013). Due to a quickly changing context, the ongoing flow of organizational life is interrupted, creating uncertainty on how to act (Maitlis, 2014). Instead of the daily stream of routines that do not demand our full attention, such a situation presents a break from the ordinary and calls for directed action (Weick et. al., 2005: 415). At the same time, what is expected and trained for seems to be insufficient as the situation somehow loses it resemblance with ‘reality’ (Maitlis, 2005). It is at this point that sensemaking sets in. Sensemaking is

13 a complex cognitive process through which individuals develop an understanding of an unfamiliar set of conditions (Caughron et al, 2011: 3). Particularly in a turbulent and dynamic context requiring collective action, such as a crisis, sensemaking can be crucial to figure out the cause of the problem, its likely outcome and the possibilities for influencing the developing situation (Weick, 1993). In doing so, individuals attempt to select and interpret cues in a comprehensible and meaningful way, enabling them to act (Kraft et. al., 2015). Therefore, sensemaking essentially focuses on the process of restoring and building shared awareness out of various competing interests and perspectives (Guillherme et al., 2010: 11).

However, unexpected events do not necessarily always trigger sensemaking. Even when cues are discrepant, sensemaking can fail to occur if the organizational culture or group norms are dominant enough to mitigate against it. For sensemaking to start, the discrepancy between what is expected and what is experienced should be remarkable enough. In other words, the issues or events have to stir up profound confusion or surprise (Maitlis, 2014). At such times, individuals or groups will ‘’ask what is going on, and what they should do next’’ (Guillherme, 2010: 11.). Weick (1993) therefore points out that sensemaking is associated with understanding the cognitive filters that people use and what they tend to include and exclude. Through these filters, individuals simplify a complex situation, to prevent becoming overwhelmed by data (Combe & Carrington, 2015). In producing these simplifications, sensemaking requires a rather detailed method to process information, share it with the right people, create a dynamic picture, analyze possible scenarios and formulate specific information needs (Boin et. al., 2013: 82).

The work of Karl Weick (1995) in particular has focused on understanding this very process. The author has presented seven interrelated sensemaking characteristics that “have an effect on the willingness of people to disengage from, discard, or ‘walk away’ from their initial story and adopt a newer story that is more sensitive to the particulars of the present context” (Weick, 2001). For the purpose of this thesis, three characteristics will be discussed: extracted cues, enactment and plausibility. The reason for such a selection is that these three elements describe the choices and actions of individuals during crises rather than the overall characteristics of the process. Therefore, as the aim is to understand the involvement of leaders in sensemaking, the chosen characteristics can enlighten how and why these individuals act during unexpected, confusing and novel events (Maitlis & Christianson, 2014).

To start, Weick (1988) argues that individuals are only able to process so much information in a complex situation. As a result, some details are noticed and receive attention, while others are set aside. The cues that an individual highlights often support an interpretation of the events preferred by him or her (McKee et. al., 2008: 8). Therefore, individuals determine the focus of attention based on what they regard as relevant and agreeable in that specific context. Ultimately, a frame of reference is created in

14 which ideas are simpler and have more familiar structures (Weick 1995: 50). Secondly, enactment focuses on the idea that organizational members create their own environment. The central point being that when people act, they bring structures, events, opportunities and constrains into existence and set them in motion. As people ‘’often don’t know what the ‘appropriate action’ is until they take some action and see what happens’’ (Weick, 1988: 307), through enactment individuals are able to learn by simply doing. By acting, complex issues are transformed to simple tasks, subsequently clarifying what the problem might be. However, besides containing the crisis to a lower level, enactment can also intensify it before it has properly been understood (ibid.). In that way, after triggering events occur, quick and spontaneous reactions solve some immediate problems, while also creating new problems (Shrivastava, 1987: 24). Lastly, sensemaking is shaped by plausibility rather than accuracy. Weick (1988) argues that an accurate perception of a situation is not always feasible due to limited cognitive means or time constraints. Besides, especially in the context of a crisis, aiming for accuracy hinders action as it often means that individuals will wait until they have gathered all the facts (ibid.). Therefore, to cope with ambiguity and uncertainty, sufficient information should be gathered to create a reasonable explanation to act upon. Ultimately, by extracting cues, creating plausible explanations and enacting the environment, individuals create the ‘reality’ that determine their action.

2.4

S

Sensemaking is a fundamentally social process as organizational members are constantly interpreting their environment in and through interactions with others (Maitlis, 2005). To illustrate the diversity of this process, Maitlis (2005) formulated four distinct sensemaking strategies based on leader and stakeholders engagement (Smerek, 2006: 59). These guided, fragmented, restricted or minimal strategies are each associated with distinct characteristics and outcomes. The strategy used during a specific crisis can be determined by analyzing whether the level of control and animation is either high or low. Control is linked to the degree to which a leader is involved in influencing the sensemaking process while animation refers to the degree to which stakeholders are engaged in the sensemaking process (Webster et.al, 2016). The level of each dimension depends on the following characteristics described by Maitlis (2005; 2007; 2014):

Control

• Leader involvement: When encountering moments of uncertainty or ambiguity, leaders

‘’seek to clarify what is going on by extracting and interpreting cues from their environment, using these as the basis for a plausible account that provides order and ‘make sense’ of what has occurred, and through which they continue to enact the environment’’ (Maitlis & Christianson, 2014: 58).

15

• Formality: Sensemaking by both stakeholders and leaders tends to occur in an organized,

systematic fashion, rather than ad hoc: controlled sensemaking processes are dominated by formal committees, scheduled meetings and planned events, rather than by informal, impromptu meetings of self-organizing groups (Maitlis, 2005: 30).

• Meeting type: A third key aspect of controlled processes is that a significant amount of

sensemaking occurs in private meetings between leaders and stakeholders, rather than in more open, public forums (Maitlis, 2005: 30)

Animation

• Stakeholder involvement: When encountering moments of uncertainty or ambiguity,

stakeholders ‘’seek to clarify what is going on by extracting and interpreting cues from their environment, using these as the basis for a plausible account that provides order and ‘’make sense’’ of what has occurred, and through which they continue to enact the environment’’ (Maitlis & Christianson, 2014: 58).

• Intensity of information: A central characteristic of animation is an intense flow of

information: leaders routinely report back to their executive teams, boards and other stakeholders, and information is also regularly shared among stakeholder groups (Maitlis, 2005: 31).

• Continuity of information: Sensemaking in animated processes tends to occur in iterative

discussions that continue over many months, as numerous stakeholders volunteer their opinions and state their demands, and leaders work to articulate their own accounts of the issues of concern (Maitlis, 2005: 31).

Maitlis (2005) argues that the dimensions control and animation can provide an important and missing component in sensemaking research as it proposes a common language with which a variety of everyday sensemaking processes can be described, contrasted and compared (Maitlis, 2005: 44). Although Maitlis’ (2005: 45) research concluded that guided sensemaking can be regarded as the most effective strategy in times of organizational change, the author emphasizes that the most optimal method may vary with the kind of outcome sought. As such, she states that researchers should examine the contexts most conductive to each one.

2.4.1GUIDED SENSEMAKING STRATEGY

Processes that are highly controlled and highly animated are regarded as guided sensemaking. In such instances, leaders are active as sensemakers, as they construct and promote explanations for the events (Maitlis, 2005). Simultaneously, stakeholders are also actively engaged in sensemaking, as they attempt to shape the beliefs about certain elements of the situation and their corresponding significance. Such

16 sensemaking processes are guided as leaders, through the confidential and systematic approach they bring to their stakeholder interactions, are able to collect, coordinate, and shape stakeholder contributions (Maitlis, 2005: 35). In her research on sensemaking in various provincial orchestras, Maitlis (2005) concluded that guided sensemaking produced a participatory and energetic process as leaders used a confidential approach to gather, coordinate and shape stakeholders contributions on the topic of income generation. At the same time, stakeholders also developed their own accounts of the situation by speaking openly on the orchestra’s financial needs and the potential role of funders. Ultimately, this produced rich and unitary accounts that captured a common understanding of the situation and the goals (Maitlis, 2005: 36) and resulted in various joint marketing, education and fund-raising initiatives.

Expectation 1: A guided sensemaking strategy will lead to vigilant decision-making process

2.4.2FRAGMENTED SENSEMAKING STRATEGY

When processes are animated but not controlled, the sensemaking strategy can be regarded as fragmented. Stakeholders animate the process by raising issues, developing and shaping accounts of the situations, and arguing for a potential solution (Maitlis, 2005: 36). Leaders, on the other hand, while seeking the views of stakeholders, do not attempt to control or organize the discussions, nor do they integrate stakeholder accounts into coherent and collective accounts (ibid.). The outcome of such a sensemaking process often consist of multiple and individualist accounts (Smerek, 2006). In their case study research on sensemaking, Watkins et al. (2017) found that fragmented sensemaking was pervasive in understanding cyber security threats in the newsrooms. The authors conducted in-depth interviews with journalists about computer security in their professional work and found that, in the vacuum of sensemaking efforts on the part of the organization, journalists actively engaged in ad hoc and individualized judgements. As a result, asymmetrical and scattered attempts were made to clarify chaotic situations. Such a strategy was regarded as a hazardous approach for security in a networked organization, as the ad hoc strategy lead to a spread of misinformation, perpetuating a divided and confusing security landscape (Watkins, 2017: 7). Ultimately, these misaligned responses to various cyber-attacks, the authors argue, can create dangerous vulnerabilities.

Expectation 2: A fragmented sensemaking strategy will lead to hypervigilant decision-making process

2.4.3RESTRICTED SENSEMAKING STRATEGY

The third strategy for sensemaking occurs when processes are highly controlled but not animated. Leaders engage in high level sensemaking by proclaiming an overarching account of the encountered issues. In doing so, they identify a few stakeholders as valuable to the process and seek them out at key points to learn from their constructions of the issues. These stakeholders are regarded as ‘social

17 anchors’. The outcome of such a restricted form is often a narrow yet unitary account with one dominant interpretation of the situation (Maitlis, 2005). A restricted sensemaking strategy was found by Kraus and Strömsten (2012) in their case study of four companies that went public on the Swedish stock exchange. The executives of all four companies underlined how hard it was to be the newcomer at the stock exchange. Therefore, faced with an uncertain situation, top managers needed to extract concrete cues to initiate their actions. To do so, a restricted sensemaking strategy was applied in which the advice of financial analysts was sought on various issues. These formal interactions with social anchors not only gave managers a feeling of what was important for the financial analysts, but it also gave them the opportunity to easily communicate with them based on a common, quantitative language understood by both parties (Kraus and Strömsten, 2012: 197). Both executives and analysts emphasized that, in the initial stage, such unitary accounts organized activities in a clearly defined way.

Expectation 3: A restricted sensemaking strategy will lead to vigilant decision-making process

2.4.4MINIMAL SENSEMAKING STRATEGY

The final sensemaking strategy is characterized by a process that is neither controlled nor animated. Animation is low as few stakeholders discuss the issue or seek to share their construction of it. Also, leaders make little attempt to promote their interpretations or to collect views of their stakeholders in a systematic way. As a result, each party tends to await the other’s interpretation (Maitlis, 2005: 42). Ultimately, this type of sensemaking is said to produce little synthesis of perspectives (Smerek, 2006: 49). A form of minimal sensemaking was observed by Stansfield and Hartley (2015) in their research on leadership in an inquiry committee of British parliament. While investigating the sensemaking process between the chair of the committee, committee members and witnesses, the researchers stated that all parties exhibited little engagement. The chair failed to engage members to develop a common view on the situation as he did not provide a common narrative. This was explained by the fact that the inquiry did not cover a controversial issue connected to the government’s main policy agenda. Therefore, the issue was regarded as having no political prominence. At the same time, stakeholders did not engage fully in the issues as they did not understand why the inquiry had been instigated in the first place. The stakeholder arena was small and the general level of energy was rather low, illustrated by members failing to make eye contact with witnesses or providing any narrative for the issue.

Expectation 4: A minimal sensemaking strategy will lead to hypervigilant decision-making process

2.5

C

:

-

Based on the discussed literature, the prediction is that guided and restricted sensemaking will have a positive effect on the vigilance of the decision-making process. Three arguments can be formulated for such an expectation. To start, the involvement of leaders can help organizations to clarify an unfamiliar

18 situation more quickly. As stated, vigilant decision-making requires a thorough understanding of the problem situation. For such an understanding, organizations have to constantly and properly assess the changing conditions. However, to grasp the problem, organizations first have to recognize that something out of the ordinary is even occurring. As stated, even when cues are discrepant, sensemaking can fail if the organizational culture or group norms are dominant enough to mitigate against it. In such instances, signs of an emerging threat can be overlooked, causing a looming crisis. Here, leaders can play a crucial role. Through their expertise and position within the organization, leaders have the ability to recognize noteworthy deviation from regular processes and subsequently bring these to the attention of others (Boin et al., 2013: 82). Their evocative language and their construction of narratives can stimulate sensemaking by emphasizing threatening cues. At the same time, by training sensemaking skills within the organization, leaders construct an atmosphere of dialogue and critical analysis, in which employees are able and willing to signal, in their eyes, deviating cues. This will benefit the decision-making vigilance of the organization as acting on early warning signs can decrease the time pressure on the decision-making process. Consequently, organizations will have additional time to consider alternatives, costs and risks, expert judgements and possible implementation methods.

Secondly, to decide on the proper response to an uncertain situation, it is necessary to view events from a variety of perspectives. However, as previously argued, this can be a quite time-consuming process as individuals can become overwhelmed by the overload of information during a crisis (Wolbers & Boersma, 2013). As a result, in organizations, threats can often lead to the consideration of fewer external cues and the reliance of tried modes of operation (Ancona, 2012: 12). Janis and Mann (1977 as cited in Kyle, 1980: 6) also point out this limitation, arguing that poor decision-making often stems from ‘’the limited nature of man’s capacity to process information’’. However, through a high level of control in the sensemaking process, leaders can prevent such information paralysis, as they will be actively constructing, singling out and rearranging novel signals. Consequently, leaders can ensure that team members will be working from a common sensemaking frame which explains ‘’what’s going on out there’’ (Ancona, 2012: 6). This will not only make the environment more comprehensible, it will give also organizational members a clear view of which information is valuable. As such, from the overload of information, relevant information can be filtered to formulate various alternatives.

Lastly, besides providing guidance to filter information, a high level of leader involvement can also stimulate coordinated action. Basically, what sensemaking accomplishes through enactment is changing individuals from inactive observers to active agents. By promoting action, leaders can enable their organizations to grasp the new environment as they instantly and constantly probe for new signs. Since a common sensemaking map is available, the need for constant guidance by leaders can be replaced with independent, yet coordinated, action. Here, the leader is tasked with encouraging rather than supervising. Especially in an era in which management by command-and-control has shifted to

19 management by teamwork and collaboration (Ancona, 2013), promoting enactment, for instance through leading by example, can be important to inspire individuals to take action. Ultimately, through enactment leaders can stimulate their organization to gather a larger number of relevant information and wider variety of alternatives to integrate this into a coherent view of the situation (Caughron et al., 2011: 4).

Figure 2. Causal mechanism: the relationship between sensemaking strategy and decision-making vigilance

To sum, it is argued that when leaders have a high level of involvement in the sensemaking process, as is the case in a restricted or guided strategy, the decision-making process will also be benefited due to three stimulating factors: additional time, a variety of perspectives and coordinated action. Subsequently, when a leaders’ involvement is low, which is the case in fragmented and minimal organizational sensemaking, it is expected that the vigilance of the decision-making process will also be low, causing hypervigilance.

20

3.

M

The main goal of this study is to determine the relationship between the sensemaking strategy and the vigilance of the decision-making process. In order to achieve this, a comparative case study design will be conducted. The methods and techniques used to gather data as well as the argumentation for these choices will be discussed in this chapter.

3.1

R

By conducting a case study, the aim is to place social processes in their own context to understand their meaning (Esterberg, 2002). To be more specific, sensemaking strategies and decision-making vigilance are researched in the crisis context through which they are shaped. Such a design is preferable as case studies are regarded as more suitable to extreme or critical situations of which little is known. Yin (2003: 1) for instance argues that ‘’case studies are the preferred strategy when “how” or “why” questions are being posed, when the investigator has little control over events, and when the focus is on contemporary phenomenon within some real-life context”. This is pre-eminently case in this research as it is will be difficult to undertake an experimental design to simulate an organizational crisis. As such, the relevant behaviors, in this case sensemaking strategies, cannot be controlled and manipulated (Yin, 2003). Besides, the careful examination of a small number of cases has the advantage that more in-depth information can be gathered on each case. This is especially useful as crises are often complex phenomena which require a thorough analysis to understand their distinct and unique characteristics.

3.2

D

Empirical data will be collected through face-to-face semi-structured interviews. The argument for such a technique is that it allows participants to give a detailed description of a complex issue, in this case sensemaking and decision-making during a crisis, while also providing enough flexibility to elaborate on thoughts and actions (Bowling, 2002: 260). Besides, a semi-structured interviewing technique is also fitting because it gives researchers the opportunity to pause, prompt and probe fittingly (Ritchie & Lewis, 2003: 141). This is especially important as, during the interview, the respondents will have the opportunity to choose the crisis they would like to discuss. Therefore, although questions will be prepared ahead of time, I will have the ability to discuss the topics that are relevant, supplemented with probing questions to learn as much as possible about the case.

3.3

C

A comparative case study design requires the synthesis of differences, similarities and patterns across two or more case that share a common goal or focus (Goodrick, 2014: 1). As such, it important to carefully select the cases under review. To be consistent in the choice of cases, a most similar case selection technique will be used. Cases have been chosen based on the type of organization as well as

21 the type of crises that they have encountered. To start, although the selected organizations operate in various industries, they are similar in magnitude. All four companies are organizations with a multimillion revenue, a large employee base and multiple branches all over the world. Secondly, all four organizations have encountered a similar type of crisis as they all fit the description of an organizational crisis. In line with Hermann’s (1972: 33) description of such a crisis, the various events threatened the high priority goals of the organizations, restricted the amount of time available for response, and surprised decision-makers by their occurrence, thereby causing high-levels of stress. The characteristics of the organizations are described in Table 2.

Organization Revenue (2017) Number of employees (2017)

Type of crisis

Organization A > EUR 20 billion > 100.000 Organizational crisis

Organization B > EUR 10 billion > 100.000 Organizational crisis

Organization C > EUR 30 billion > 200.000 Organizational crisis

Organization D > EUR 10 billion > 50.000 Organizational crisis

Table 2. Characteristics of the organizations

Furthermore, as illustrated in Figure 3, the unit of analysis will be crisis managers in the private sector. The argument for such a choice is that it zooms in on the views and action of those that are immediately and directly involved in the crisis response. Crisis managers form a group of professionals who specifically assist organizations in preparing for, responding to and recovering from a crisis that can pose a fundamental threat to their functioning and goals. Such threats include but are not limited to severe consequences for the day-to-day business operations, reputation, financial resources and morale of the organization. By researching this group, new insights are gathered on the research topic, sensemaking and decision-making in times of crises, from those that have the required experience and expertise. Due to this narrow unit of analysis, non-probability sampling will be used. This means that respondents are deliberately selected for the valuable information they provide that cannot be gotten from other choices (Maxwell, 1997: 87).

22

………...

Figure 3. Research scope

3.4OPERATIONALIZATION

This study will have a deductive approach as pre-determined indicators will be used as a guiding framework. Based on the combination of the indicators animation and control, the independent variable ‘sensemaking strategy’ will be identified as either guided, restricted, fragmented or minimal. The assessment of the indicators is based on a careful and detailed approach to interviewing, stimulating the respondents to describe the processes in which they were or weren’t engaged, and following up with questions that allow me to categorize their descriptions. In this, Maitlis’ (2005) detailed account of the characteristics of an animated and controlled sensemaking strategy will also be used as a guideline. Also, the three selected properties of Weick (1995), which describe the characteristics of sensemaking, will be used to determine the involvement of leaders and stakeholders. Subsequently, the dependent variable is assessed based on seven criteria for vigilant decision-making, described by Janis and Mann (1977). As the authors used quantitative data in their initial research, the assessment in this study will rely on later qualitative work conducted by Herek, Janis and Huth (1987). In their analysis, the decision-making process is regarded as vigilant when the organization is able to successfully meet at least six criteria. Hypervigilance occurs when the decision-makers fail to meet four or more criteria. When organizations fail to meet two or three criteria, they are neither vigilant nor hypervigilant (Janis et. al., 1987).

Assessment of decision-making process Number of completed criteria

Vigilant 6 or more

Hypervigilant 4 or 5

Neither vigilant or hypervigilant 3 or less

Table 3. Assessment decision-making vigilance

To prevent that biased questions influence respondents’ answers, the interview will start with the following general question:

Context: Organizational crises

Unit of Analysis: Crisis managers in the private sector

Unit of observation: Crisis manager organization A Unit of observation: Crisis manager organization B Unit of observation: Crisis manager organization C Unit of observation: Crisis manager organization D

23 Could you describe, in detail and chronological order, which steps you and your team took after you were informed of the threatening situation?

Subsequently, when the absence of a criterion is suspected, probing questions will be asked. The operationalization of the probing questions is listed in Table 4.

Variable Concept Indicator Measurement Description Independent variable: Sensemaking strategy Guided sensemaking

high control Crisis managers

are highly involved in the sensemaking process

• Crisis managers

extract and interpret cues • Crisis managers develop plausible accounts to make sense of what is occurring • Crisis managers enact the environment Sensemaking occurs in a systematic organized fashion • sensemaking processes are dominated by formal committees, scheduled meetings and planned events A significant amount of sensemaking occurs in private meetings • processes in which there tends to be numerous private and one-on-one interactions between leaders and stakeholders

high animation Stakeholders are

highly involved in the sensemaking process

• Stakeholders

extract and interpret cues • Stakeholders develop plausible accounts to make sense of what is occurring • Stakeholders enact the environment

24 Sensemaking occurs through an intense flow of information • leaders routinely reported back to their boards, executive teams, and other stakeholders, and information is also regularly shared among stakeholder groups Sensemaking has a continuous and iterative rhythm • Sensemaking occurs in iterative discussions that continued over an extended period, as numerous stakeholders volunteer their opinions and state their demands, and leaders work to articulate their own accounts of the issue

Restricted sensemaking

high control (see guided

sensemaking)

(see guided sensemaking)

low animation Stakeholders are

not involved in the sensemaking process

• Stakeholders do not

actively extract and interpret cues • Stakeholders do not actively develop plausible accounts to make sense of what is occurring • Stakeholders do not

actively enact the environment Sensemaking occurs through a light flow of information • Few stakeholders are aware of an issue's existence or importance, and still fewer knowing its details.

25 Sensemaking has a broken and intermittent rhythm • Sensemaking primarily occurred through sporadic leader-stakeholder discussions that took place over an extended period. Fragmented

sensemaking

low control Crisis managers

are not involved in the sensemaking process

• Crisis managers do

not actively extract and interpret cues

• Crisis managers do not actively develop plausible accounts to make sense of what is occurring • Crisis managers do

not actively enact the environment Fewer mechanisms organize sensemaking activities • Sensemaking processes are dominated by informal, impromptu meetings of self-organizing groups Sensemaking takes place in open forums that include a large number and variety of parties

high animation (see guided

sensemaking)

(see guided sensemaking)

Minimal sensemaking

low control (see fragmented

sensemaking)

(see fragmented sensemaking)

low animation (see restricted

sensemaking)

(see restricted sensemaking)

26 Dependent variable: Decision-making vigilance Stage 1: Objectives 1. Decision-makers appraise the situation and survey the full range of objectives to be fulfilled;

Did you and your team

collaboratively agree of the severity situation? Did you and your team survey the full range of objectives to be fulfilled? If so, what were the objectives? Stage 2: Alternatives 2. Decision-makers thoroughly canvas a wide range of alternative courses of action

Did you and your team consider a wide range of alternative courses of actions during the crisis? If so, what were the alternatives? Stage 2: Consequences 3. Decision-makers carefully consider whatever is known about the costs and risks of negative consequences, as well as the positive consequences, that could flow from each alternative

Did you and your team consider the costs, risks and consequences of every alternative? If so, what were the costs, risks and consequences?

Stage 2: Information

4. Decision-makers

intensively search for new information relevant to evaluate the alternatives

Did you and your team constantly update information to learn more about every alternative? Stage 3: Assimilation and experts 5. Decision-makers correctly assimilate information and take into account any new information or expert judgement, even when the

Did you and your team assimilate the gathered data to get an overall view of the alternatives? Did you and your team gather views

27 information or

judgement does not support the course of action preferred on the situation from experts? Stage 3: Reexamination 6. Decision-makers

reexamine all the possible consequences of all known alternatives, including those originally regarded as unacceptable, before making a final choice

Did you and your team reexamine the possible consequences of all known alternatives before making a choice? If so, what was the outcome of this reassessment? Stage 4: Preparation and monitoring 7. Decision-makers make detailed preparations to implement and monitor the chosen alternative, with special attention to contingency plans that might be required if various known risks were to materialize

Did you and your team make detailed preparations to implement and monitor the chosen alternative? If so, which measures did you take?

Table 4. Operationalization independent (Maitlis, 2005: 30-31) and dependent variable (Janis & Mann, 1977: 11)

3.5DATA ANALYSIS

The gathered data will be analyzed by systematically arranging and researching interview transcripts. As Maitlis’ (2005) sensemaking strategies and Janis and Mann’s (1977) criteria for vigilant decision-making are used to extend a theoretical framework, deductive coding will be applied. A deductive approach is explicit about the applied categories as it begins with the generation of a provisional list of codes before starting fieldwork (Miles and Huberman, 1994: 121). However, while progressing with the analysis, it is still possible to revise, add, modify or delete specific provisional codes. The operationalized indicators as well as their measurements have been listed in Table 4. In accordance with such an approach, directed content analysis is employed, using predetermined codes. Having such a priori codes gives the opportunity to specifically search for certain aspects of a phenomenon, making the coding process more structured. To search for these provisional codes and subsequently assess and

28 interpret the findings, ATLAS.ti, a qualitative data analysis program, will be used. Transcript of the interview can be found in the appendix.

3.6

R

The reliability of a research is determined by the quality of its measurement process. In other words, a research is reliable when it can be repeated by others. When the exact same research is conducted under the same conditions, stable and consistent results have to be generated (Koo & Li, 2016). Reliability is often an issue when it comes to conducting interviews. During a semi-structured interview, respondents have the ability to talk freely and elaborately on specific topics of interests. Repeating an interview can therefore be quite of a challenge as not all the questions are pre-determined. Besides, answers might also change over time due to various factors such as the respondent’s personal situation and priming effects. Nevertheless, the reliability of the interview can be increased by including transcripts as well as the questions of the interview. These can be found in the Appendix A.

Next, the internal and external validity of the data collection methods needs to be assessed. Internal validity discusses the extent to which a research is actually measuring what it is supposed to be measuring (Alshenqeeti, 2014). Determining the internal validity is important in explanatory case studies as the aim is to demonstrate that a certain outcome, the vigilance of the decision-making process, was influenced by the independent variable, the sensemaking strategy (Ellram, 1996). An obstacle to reaching such internal validity is the research design. As this study relies on comparative case studies, rather than an experimental design, it will be difficult to control extraneous variables. As a result, it will not be possible to fully rule out that the found results might be caused by other factors. However, to increase the internal validity, cases and respondent will be selected that are as similar as possible. Through purposive sampling the likelihood of a causal relationship can be increased as the homogenous case selection can eliminate as many external factors as possible. The characteristics used for such a selection can be found in Table 2.

External validity on the other hand requires a research to be generalizable. When a research is generalizable, its findings from the sample are also applicable to the entire research population. In general, the external validity of case studies is regarded as low, as the unique settings and characteristics of cases make it difficult to generalize them to other contexts (Yin, 2003). Besides, due to the small number of respondents, the generalizability of the research can also be called into question. As a result of time constraint, only four respondents were selected based on non-probability sampling. However, the primary goal of this thesis is not so much to generalize the findings to other cases or a larger population (Bryman & Bell, 2007). Instead, the main purpose is to instigate a discussion on the role of crisis managers in the sensemaking process and to stimulate more research on the topic. Based on a

29 greater number of case studies examining the relationship between sensemaking strategies and decision-making processes, academics will be able to make more general statements in the future.

30

4.

A

The analysis of the gathered data will be carried out in two main stages. First, the strategy for sensemaking will be determined by analyzing how animated and controlled the process was. Subsequently, the vigilance of the decision-making process will be examined by researching the extent to which the process adhered to the seven requirements discussed in the theoretical framework. However, before presenting the results from the interviews, every case study will start with a brief description of the crisis, to provide a solid background for the analysis.

Figure 4. Process of analysis

4.1

O

A

To execute a conversion in the IT systems of organization A it was necessary to clear all stored data present at a specific location. To do so, an employee decided to make a back-up of the data and temporarily place it in a different file. However, it was overlooked that by doing so the data would be accessible to all employees of the organization. As a result, for a brief period of time, employees were able to access privacy sensitive information of their colleagues and former colleagues. To deal with the crisis a specialized crisis assessment and treatment team, led by the crisis manager, was activated. Overall, the crisis lasted for approximately one week.

Case description • Actors • Characteristics Sensemaking strategy • Level of control - Leader involvement - Formality - Meeting type • Level of animation - Stakeholder involvement - Intensity of information - Continuation of information Vigilance of decision-making process •Objectives •Alternatives •Consequences •Information •Assimilation and experts •Reexamination •Preparation, implementation and montioring

Guided strategy Restricted strategy

Fragmented strategy Minimal strategy

Vigilant

Hypervigilant